In order to make the system more robust to human orientation changes, i need Pose normalization method, but i dont know it's Related Work, or which work is beter than other work ?
I Need Help
Regards
Transforming the 3D point cloud into PCA space for subsequent features extraction could be a possible solution. Nevertheless, following papers utilizing some histogram oriented techniques can also be helpful :
(1) Suryanarayan, P., Subramanian, A. and Mandalapu, D., 2010, August. Dynamic hand pose recognition using depth data. In Pattern Recognition (ICPR), 2010 20th International Conference on (pp. 3105-3108). IEEE.
(2) Oreifej, O. and Liu, Z., 2013. Hon4d: Histogram of oriented 4d normals for activity recognition from depth sequences. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 716-723).
(3) Tang, S., Wang, X., Lv, X., Han, T.X., Keller, J., He, Z., Skubic, M. and Lao, S., 2012, November. Histogram of oriented normal vectors for object recognition with a depth sensor. In Asian conference on computer vision (pp. 525-538). Springer Berlin Heidelberg.
(4) Zhang, C. and Tian, Y., 2015. Histogram of 3D facets: A depth descriptor for human action and hand gesture recognition. Computer Vision and Image Understanding, 139, pp.29-39.
First do limb normalization as we different skeletons have different body lengths which can be done by making center of body(Sacrum) as center of co-ordinate axis
In order to get a view invariant relative position vector for every node,multiply Rotation matrix(R) with Limb normalized vectors.Rotation matrix can be found with Gram-Schmidt ortho normalization process
- Badges
- Science topic